The patience and precision involved with drawing geometric patterns in sand is right up a robot’s alley, and demonstrating this is [rob dobson]’s SandBot, a robot that draws patterns thanks to an arm with a magnetically coupled ball.
SandBot is not a cartesian XY design. An XY frame would need to be at least as big as the sand table itself, but a SCARA arm can be much more compact. Sandbot also makes heavy use of 3D printing and laser-cut acrylic pieces, with no need of an external frame.
[rob]’s writeup is chock full of excellent detail and illustrations, and makes an excellent read. His previous SandBot design is also worth checking out, as it contains all kinds of practical details like what size of ball bearing is best for drawing in fine sand (between 15 and 20 mm diameter, it turns out. Too small and motion is jerky as the ball catches on sand grains, and too large and there is noticeable lag in movement.) Design files for the SCARA SandBot are on GitHub but [rob] has handy links to everything in his writeup for easy reference.
Sand and robots (or any moving parts) aren’t exactly a natural combination, but that hasn’t stopped anyone. We’ve seen Clearwalker stride along the beach, and the Sand Drawing Robot lowers an appendage to carve out messages in the sand while rolling along.
Consider the humble ball bearing. Ubiquitous, useful, and presently annoying teachers the world over in the form of fidget spinners. One thing ball bearings aren’t is easily 3D printed. It’s hard to print a good sphere, but that doesn’t mean you can’t print your own slew bearings for fun and profit.
As [Christoph Laimer] explains, slew bearings consist of a series of cylindrical rollers alternately arranged at 90° angles around an inner and outer race, and are therefore more approachable to 3D printing. Slew bearings often find application in large, slowly rotating applications like crane platforms or the bearings between a wind turbine nacelle and tower. In the video below, [Christoph] walks us through his parametric design in Fusion 360; for those of us not well-versed in the app, it looks a little like magic. Thankfully he has provided both the CAD files and a selection of STLs for different size bearings.
[Christoph] is no stranger to complex 3D-printable designs, like his recent brushless DC motor or an older clock build. The clock is cool, but the bearings and motors really get us — we’ll need such designs to get to self-replicating machines.
Depending on whom you ask, fidgeting is an unsightly habit or a necessity for free-form ideation. Fan of the latter hypothesis? Well, why aren’t you making yourself a fidget pyramid?
[lignum] sculpted his fidget toy out of a chunk of 2000 year old bog-oak using hand tools and a little precision help from a Kuka KR 150 industrial robot arm. A push button, a toggle switch, a ball-bearing, and a smooth side provide mindless distraction on this piece.
Two plates of 1.5mm aluminium — also cut using the robot arm — are used to attach the button and toggle to the tetrahedron, while the ball bearing is pushed onto a cylindrical protrusion left during the cutting process for the purpose. The build video makes it look easy.
Most inexpensive 3D printers use a type of lead screw to move some part of the printer in the vertical direction. A motor turns a threaded rod and that causes a nut to go up or down. The printer part rides on the nut. This works well, but it is slower than other drive mechanisms (which is why you don’t often see them on the horizontal parts of a printer). Some cheap printers use common threaded rod, which is convenient, but prone to bad behavior since the rods are not always straight, the threads are subject to backlash, and the tolerances are not always the best.
More sophisticated printers use ACME threaded rod or trapezoidal threaded rods. These are made for this type of service and have thread designs that minimize things like backlash. They typically are made to more exacting standards, too. Making the nut softer than the rod (for example, brass or Delrin) is another common optimization.
However, when lead screws aren’t good enough, mechanical designers turn to ball screws. In principle, these are very similar to lead screws but instead of a nut, there is a race containing ball bearings that moves up and down the screw. The ball bearings lead to less friction.
Misumi recently posted a few blog articles about ball screws. Some of the information is basic, but it also covers preloading and friction. Plus they are promising future articles to expand on the topic. If you prefer to watch a video, you might enjoy the one below.
[Christian] wrote and sells some CAM/CNC controller software. We’re kinda sticklers for open source, and this software doesn’t seem to be, so “meh”. But what we do like is the Easter egg that comes included: the paths to mill out the base for a clock, and then the codes to move steel ball-bearings around to display the time.
Of course we’d like to see more info (more, MORE, MOAR!) but it looks easy enough to recreate. We could see redesigning this with marbles and a vacuum system, for instance. The seats for the ball bearings don’t even need to be milled out spheres. You could do this part with a drill press. Who’s going to rebuild this for their 3D printer? You just have to make sure that the machine is fast enough to move the balls around within one minute.
[Martin Raynsford] figured out a way to sneak some learning into a fun package. He did such a good job the test subjects didn’t even know they were teaching themselves just a tiny bit of CNC programming.
He showed off the rig at the Maker Faire. It takes simple commands as cardinal directions and units of movement. The ‘player’ (remember, they’re secretly learning something, not just playing a game) inputs a series of movements such as “N10,E10” which are then pushed through a serial connection to the Arduino. It follows these commands, moving the hidden magnet which drags the ball bearing along with it. It’s simple, but watch the clip after the break and we think you’ll agree the sound of the stepper motors and the movement of the ball will be like crack for young minds.
[Lior Elazary] designed and built this clock to simulate the function of a CPU. The problem is that if you don’t already have a good grasp of how a CPU works we think this clock will be hopelessly confusing. But lucky for us, we get it, and we love it!
Hour data is shown as a binary number on Register A. This is the center column of red parts and is organized with the MSB on the bottom, the LSB on the top, and left-pointing bits function as digital 1. The clock lacks the complexity necessary for displaying any other time data. But that’s okay, because the sound made by the ball-bearing dropping every minute might drive you a bit loony anyway. [Lior] doesn’t talk about the mechanism that transports that ball bearing, but you can see from the video after the break that a magnet on a circular path picks it up and transports it to the top of the clock where gravity is used to feed the registers. There are two tracks which allow the ball to bypass the A register and enter the B register to the right. This works in conjunction with register C (on the left) to reset the hours when the count is greater than 11.